fundamental advancements in pre-chamber spark ignition … zigler_2019_o_4.30_4.02pm_jl...chamber...

1

Fundamental Advancements in Pre-Chamber Spark Ignition and Emissions Control for Natural Gas Engines

Presenter: Brad Zigler (NREL)

Doug Longman (ANL)Brad Zigler (NREL)Scott Curran (ORNL)Mark Musculus (SNL)

June 13, 2019

DOE Vehicle Technologies Program 2019 Annual Merit Review and Peer Evaluation Meeting

This presentation does not contain any proprietary, confidential, or otherwise restricted information.

Project ID # FT080

2

Overview

• Project start date: 10/1/2017• Project end date: 12/31/2019• Percent complete: ~66%

Budget• Total project funding: $3M

– DOE share: $3M– Contractor share: $0

• Funding from FY 2018– Equally split amongst

ANL/NREL/ORNL/SNL at $750K each for total project

• Natural gas engines need to improve dilution tolerance and lean operation to achieve diesel-like efficiency

• Fundamental understanding (physics, thermodynamics, and chemistry) is necessary for improving natural gas combustion efficiency

• Fundamental catalysis research for methane conversion is needed due to challenge of methane activation

Timeline Barriers / Technical Targets

• ANL / NREL / ORNL / SNL• Industry collaboration:

– Altronic– Analytik-Service Gesellschaft (ASG)

mbH– MAHLE– Daimler Trucks North America (Detroit

Diesel)

Partners

ANL: Argonne National LaboratoryNREL: National Renewable Energy LaboratoryORNL: Oak Ridge National LaboratorySNL: Sandia National Laboratories

3

Relevance

• DOE Vehicle Technologies Office (VTO) has specific input regarding natural gas (NG) engine research needs for efficiency and emissions– Annual Natural Gas Vehicle Technology Forum– Natural Gas Vehicle Research Workshop (July 2017), which fed VTO’s funding

opportunity announcement (FOA) and the Lab Call that resulted in this multi-lab project

• Key high-level NG engine research needs:– Research needed to address barriers for achieving diesel like efficiency for NG engines – Ignition technology to enable ultra-lean operation (pre-chamber, volumetric ignition)– Fundamentals for improving NG combustion efficiency (physics, thermodynamics and

chemistry)– Low temperature combustion (LTC) concepts conceivable for NG engines, ensure real-

world mode switching and emissions control compatibility– Advances in computational fluid dynamics (CFD) and modeling for NG engines– Avoiding knock and abnormal combustion (i.e. low speed pre-ignition)– Fundamental catalysis research for methane conversion is needed due to challenge of

methane activation– Research needed for both stoichiometric and lean engine (LTC and conventional)

emission control

4

Relevance

This project focuses on early stage research focusing on pre-chamber spark-ignition (PCSI) to achieve diesel-like efficiency in medium duty (MD) and heavy duty (HD) NG gas engines by extending the lean dilution limit and/or exhaust gas recirculation (EGR) dilution limit, as well as shortening burn duration, with integrated aftertreatment

Impact:This project integrates experimental and simulation based tasks to address four key barriers to market penetration of PCSI for MD/HD NG engines:Barrier 1 – Inadequate science base and simulation tools to describe/predict the fluid-mechanical and chemical-kinetic processes governing PCSI to enable engineers in industry to optimize designs for efficiency, noise, reliability, pollutant formation, emissions control integration, and drivabilityBarrier 2 – Limited ability to extend EGR and/or lean dilution limits at higher loadsBarrier 3 – Increased propensity for PCSI hot-spot pre-ignition at high loads relative to spark ignitionBarrier 4 – Ineffective methane catalysts for the high engine-out unburned fuel concentrations coupled with low exhaust temperatures (<<400 °C) of high efficiency engines

5

Approach

Chamber Sim. (NREL)

Engine Sim. (ANL)

Metal Engine Exp’t. (ANL)

Optical Engine Exp’t. (SNL)

Bench Scale

Catalyst Exp’t. (ORNL)

Metal Engine Exp’t. (ORNL)

Collaboration and integration across four national labs connect fundamental experiments and modeling to practical hardware

Single Cylinder Multi Cylinder

Validation

Complementary Insight Fundamental Insight

Target Conditions

Complementary

Insight

Target Conditions

DOE laboratory expertise and capabilities focus on early-stage research to address key barriers for NATURAL GAS engines

Chamber Exp’t. (NREL)

6

Approach

Bench Scale Single Cylinder Multi CylinderSimulationEngine Sim. (ANL)High fidelity CFD focusing on PCSI output mixing, with ignition and flame propagation models.

Chamber Sim. (NREL)Zero dimensional (0D) and CFD map PCSI composition output to explore main chamber ignition sensitivity.

Chamber Exp’t. (NREL)PCSI added to constant volume chamber to study pre-chamber variable effects on main chamber ignition over lean / dilute conditions.

Catalyst Exp’t. (ORNL)Synthesize novel methane oxidation catalysts (MOCs) and evaluate performance with PCSI NG engine exhaust conditions.

Metal Engine Exp’t. (ANL)Single cylinder engine experiments with borescope access to study PCSI effects on lean / dilute operation and efficiency / engine-out emissions tradeoffs.

Optical Engine Exp’t. (SNL)Single cylinder engine experiments to study PCSI output penetration to main chamber and characterize flame propagation vs. sequential autoignition.

Metal Engine Exp’t. (ORNL)Modified HD engine with PCSI in all cylinders to study dilution tolerance, conduct thermodynamic analysis of efficiency potential tradeoffs related to lean / dilute combustion with PCSI, and provide exhaust information for MOC studies.

Bench Scale Multi CylinderNREL

Chamber Exp’t.NREL

Chamber Sim.ORNL

Catalyst Exp’t.

Single CylinderANL

Engine Sim.ANL

Metal EngineSNL

Optical EngineORNL

Metal Engine

Modular PCSI designs with as much commonality as possible are used across all platforms

7

Date Description of Milestone Status Lab

March 2018 Select synthesis strategy for improved MOC performance Complete ORNL

Sept. 2018 Single cylinder research engine configuration completed Complete ANL

Sept. 2018 PSCI combustion model evaluation to define what models are suitable for PCSI simulation

Complete ANL

Dec. 2018 Initial in-cylinder ignition combustion optical imaging Complete SNL

March 2019 PCSI scoping experiments in Advanced Fuel Ignition Delay Analyzer (AFIDA)

Complete NREL

March 2019 Single-cylinder data collection and post-processing Complete ANL

June 2019 Initial sensitivity analysis completed using 0D or CFD simulations based on PCSI AFIDA experiments

On-track NREL

Sept. 2019 Complete multi-cylinder engine dilution tolerance studies On-track ORNL

Sept. 2019 Presentation describing progress toward a conceptual model description of in-cylinder PCSI processes

On-track SNL

Dec. 2019 Combustion model validation against experimental engine and optical data

On-track ANL

Approach - Milestones

A subset of key milestones are presented. The ANL / NREL / ORNL / SNL team meet semi-monthly to coordinate research tasks and share results. VTO updates are also coordinated.

8

△

Technical Accomplishments and Progress

• NREL modified the AFIDA for studies of PCSI and main chamber pressure rise and ignition progress with well controlled temperatures and pressures, sweeping equivalence ratio and simulated EGR rates, providing guidance for SNL optical and ANL + ORNL metal engine experiments

• 0D and CFD simulations map variations in PCSI jet composition to variations in flame speeds and ignition propensity, producing sensitivity factors for main chamber ignition quality in terms of mixture composition

• Initial results indicate OH, CH2O, O, & H radical pool output in jets are most critical to ignitionBench Scale Multi Cylinder

NRELChamber Exp’t.

NRELChamber Sim.

ORNLCatalyst Exp’t.

Single CylinderANL

Engine Sim.ANL

Metal EngineSNL

Optical EngineORNL

Metal Engine

Constant volume PCSI experiments and 0D + CFD simulations provide sensitivity studies

Spark

Dashed line = pre-chamberSolid line = main chamber

𝜆𝜆 = Lambda, air-fuel equivalence ratioms = Milliseconds

𝝀𝝀 = 1.0

𝝀𝝀 = 1.2

𝝀𝝀 = 1.4𝝀𝝀 = 1.6

OH: Hydroxyl radical CH2O: Formaldehyde O: Oxygen (elemental) H: Hydrogen (elemental)

B1: Science base

B2: Dilution

9


• ANL metal single cylinder engine studies demonstrated passive PCSI fueling extended lean limit to lambda = 1.6 (with same combustion duration / stability of SI)

• Active PCSI fueling significantly extended the lean flammability limit and enabled stable combustion at lambda > 2.2 by leveraging fuel-rich mixture inside the pre-chamber

• Lean limit extension beyond lambda ~ 1.8 required fuel-rich mixture inside the pre-chamber, but strength of rich mixture had no observed influence within the flammability limits

• Results suggest fuel-rich pre-chamber produces chemically active jets that ignite lean chargeBench Scale Multi Cylinder

NRELChamber Exp’t.

NRELChamber Sim.

ORNLCatalyst Exp’t.

Single CylinderANL

Engine Sim.ANL

Metal EngineSNL

Optical EngineORNL

Metal Engine

Engine experiments show rich pre-chamber extends lean limits, with chemically active jetsbTDC: Before top dead center

CAD: Crank angle degrees

IMEPn: Net indicated mean effective pressure

PCI: Pre-chamber ignition

RPM: Revolutions per minute

SI: Spark ignition

B1: Science base

B2: Dilution

10


• ANL simulations in CONVERGE showed G-Equation has higher potential to capture pre-chamber combustion events compared to other models (multi-zone well-stirred reactor; extended coherent flame model)

• G-Equation can account for both large and scale turbulences found in the pre-chamber, and these factors were tuned to match experiments at two operating conditions

• Strong turbulence-chemistry interaction is expected when jets exit pre-chamber (~ 1% MFB), and flame propagation dominates at very late stages



Chamber Sim.ORNL

Catalyst Exp’t.

Single CylinderANL

Engine Sim.ANL

Metal EngineSNL

Optical EngineORNL

Metal Engine

𝝀𝝀 = 𝟏𝟏.𝟔𝟔𝟔𝟔

CFD 5 cyclesEXP 300 cycles

1200 rpm7.5 bar IMEPg

CA: Crank angleDa: Damkohler number

deg: DegreesIMEPg: Gross indicated mean effective pressure MFB: Mass fraction burned

Engine simulations developed to accurately predict PCSI NG engine combustion

J: JouleKa: Karlovitz number MPa: Megapascals s: Seconds

B1: Science base

B2: Dilution

11


• Initial studies with passively fueled pre-chamber reveal large cycle-to-cycle variation in timing of individual pre-chamber jet ignition even when COV of IMEP is low

• At leaner conditions, pre-chamber jet luminosity becomes increasingly variable, and main chamber combustion progression slows and becomes less flame-like and more distributed

• SNL experiments validate/inform NREL constant volume chamber and ANL SCE studies / simulations, and support development of a conceptual-model description of PCSI



Chamber Sim.ORNL

Catalyst Exp’t.

Single CylinderANL

Engine Sim.ANL

Metal EngineSNL

Optical EngineORNL

Metal Engine

Optical engine studies provide fundamental NG mixing and combustion data with PCSI

Visible combustion luminosity (broadband chemiluminescence, no filtering)Lambda = 1.5 Lambda = 1.7 Lambda = 1.9

COV: Coefficient of variation IMEP: Indicated mean effective pressure SCE: Single cylinder engine

B3: Pre-ignition

B1: Science base

B2: Dilution

12




Chamber Sim.ORNL

Catalyst Exp’t.

Single CylinderANL

Engine Sim.ANL

Metal EngineSNL

Optical EngineORNL

Metal Engine

Optical engine studies provide fundamental NG mixing and combustion data with PCSI

AHRR: Apparent heat release rate kPa: Kilopascals

Lambda = 1.5 Lambda = 1.7 Lambda = 1.9

• Initial studies with passively fueled pre-chamber reveal large cycle-to-cycle variation in timing of individual pre-chamber jet ignition even when COV of IMEP is low

• At leaner conditions, pre-chamber jet luminosity becomes increasingly variable, and main chamber combustion progression slows and becomes less flame-like and more distributed

• SNL experiments validate/inform NREL constant volume chamber and ANL SCE studies / simulations, and support development of a conceptual-model description of PCSI

B3: Pre-ignition

B1: Science base

B2: Dilution

13




Chamber Sim.ORNL

Catalyst Exp’t.

Single CylinderANL

Engine Sim.ANL

Metal EngineSNL

Optical EngineORNL

Metal Engine

B3: Pre-ignition

B1: Science base

B2: Dilution

• Focused dilution tolerance studies will link with single-cylinder studies and simulations • 1st and 2nd law studies will provide insight on how PCSI shifts thermodynamic balances and

to understand what additional opportunities for improved efficiency exist• Will provide exhaust composition data to MOC study

V15 N6-12 V15 N6-15 V30 N6-15 V60 N8-13

Volume 1.5 cc 1.5 cc 3.0 cc 6.0 cc

# of nozzle holes 6 6 6 8

Nozzle hole diam. 1.2 mm 1.5 mm 1.5 mm 1.3 mm

Pre Chamber Bodies for MCE DD13 Diesel DD13

• ORNL adapted a prototype modular MAHLE PCSI design to the DD13… a robust system with engineering support was necessary, while still allowing links to ANL metal and SNL optical single cylinder engine studies, and ANL simulations

Prototype MAHLE PCSI modules

V15 N6-12V15 N6-12 V15 N6-15 V30 N6-15

V60 N8-13

PCSI adapted multi-cylinder engine enables dilution tolerance and thermodynamic studies

MCE: Multi-cylinder engine

14


Developing new Methane Oxidation Catalyst (MOC) for low temperature CH4 conversion



Chamber Sim.ORNL

Catalyst Exp’t.

Single CylinderANL

Engine Sim.ANL

Metal EngineSNL

Optical EngineORNL

Metal Engine

ACEC light-off protocol strategy

Lean-MOC

[200 Lflow/(gcat*h)]

H2O 12%

O2 9%

CO2 6%

CH4 3000 ppm

CO 2000 ppm

NO 500 ppm

Ar Balance

Al: AluminumAr: Argoncat: CatalystCH4: MethaneCO: Carbon monoxideCO2: Carbon dioxideg: Gramh: HourH: Hydrogen (chemical element)H2O: WaterL: LiterMg: Magnesium

min: MinuteNH4: AmmoniumNO: Nitric oxideO: Oxygen (chemical element)O2: Oxygen (molecular allotrope)Pd: Palladiumppm: Parts per millionSi: SiliconSSZ-13: Aluminosilicate zeolite

mineral possessing 0.38 × 0.38 nm micropores

Definitions for this slide

B4: CH4 catalysts

Synthetic exhaust composition

15


Developing new Methane Oxidation Catalyst (MOC) for low temperature CH4 conversion



Chamber Sim.ORNL

Catalyst Exp’t.

Single CylinderANL

Engine Sim.ANL

Metal EngineSNL

Optical EngineORNL

Metal Engine

ACEC light-off protocol strategy

Lean-MOC

[200 Lflow/(gcat*h)]

H2O 12%

O2 9%

CO2 6%

CH4 3000 ppm

CO 2000 ppm

NO 500 ppm

Ar Balance

Synthetic exhaust composition

B4: CH4 catalysts

Modified base Si-O chabazite cage structure from (Martin, N.; Moliner, M.; Corm, A. Chem. Commun., 2015, 15, 9965)

Approach: • ORNL synthesized a series of catalysts to lower light-off

temperature of methane (CH4) oxidation – modifying the Pd active site to promote H abstraction using Mg

Accomplishments:• Completed synthesis of Pd/SSZ-13 and Mg /SSZ-13• Examined multiple calcination and hydrothermal treatments• Evaluated MOCs on a gas flow reactor using a synthetic

exhaust flow for a lean natural gas engine – Followed U.S. DRIVE (ACEC) catalyst protocol

16

Responses to Previous Year Reviewers’ Comments

This project has not previously been reviewed at a VTO Annual Merit Review.

17

Collaboration and Coordination

• ANL• Doug Longman (PI)• Riccardo Scarcelli• Sibendu Som• Ashish Shah• Joohan Kim• Munidhar Biruduganti

• ORNL• Scott Curran (PI)• Josh Pihl• Jim Szybist• Melanie DeBusk• Sreshtha Sinha Majumdar

• NREL• Brad Zigler (PI)• Matt Ratcliff• Mohammad Rahimi (post-doc)• Shashank Yellapantula• Whitney Collins• Jon Luecke• Ray Grout

• SNL• Mark Musculus (PI)• Zheming Li (post-doc)• Rajavasanth Rajasegar (post-doc)• Yoichi Niki (visiting scientist)• Dalton Carpenter (2018 intern)

• ANL / NREL / ORNL / SNL collaboration – Integrated team of leading experts– Hold semi-monthly research coordination and data exchange meetings

18

Collaboration and Coordination

• Altronic– Supplied NGI-1000 flexible natural gas engine spark ignition

system to all four DOE labs to support experiments• ASG Analytik-Service Gesellschaft mbH

– Integrated revised controls and data acquisition for PCSI module in NREL’s Advanced Fuel Ignition Delay Analyzer (AFIDA)

• MAHLE– Collaborated with ORNL to integrate MAHLE Turbulent Jet

Ignition (TJI) PCSI system for DD13 multi-cylinder engine experiments

• Daimler Trucks North America (Detroit Diesel) – Collaborated with ORNL to provide details for modification and

support for DD13 for multi-cylinder engine experiments

19

Remaining Challenges and Barriers

While the ANL, NREL, ORNL, and SNL research tasks are highly collaborative and integrated, they are still low TRL in nature…

Additional research and development is necessary for industry to commercialize high efficiency NG engine based on PCSI.

B1: Science base

B2: Dilution

B3: Pre-ignition

B4: CH4 catalysts

• We are developing a fundamental science base and simulation tools to predict fluid-mechanical and chemical-kinetic processes governing PCSI

• Our conclusions will apply generally to design of PCSI for highly dilute / lean combustion, rather than to specific hardware / strategy optimization

• Although insight will be gained, fully addressing pre-ignition at high loads is outside the scope

• We will have bench-scale MOC research, but not full catalyst development or engine integration

20

Proposed Future Research

For the remainder of this funded project:• Continued sensitivity studies on PCSI geometry parameters,

pre-chamber mixture strength, and ignition performance over a range of EGR and lean main chamber dilution

• CFD simulations and optical engine studies to provide conceptual insight into the ignition processes and characteristics of the turbulent flames / jets

• Integration of the single cylinder studies to multi-cylinder dilution / thermodynamics studies

• Feedback of exhaust characterization to MOC studiesWith additional funding, proposed future research*:• Integrate newly developed science base and tools to guide PCSI

research in conjunction with lean / EGR dilute engine combustion development and new MOC for high efficiency engine demonstration – potential industry partnership

* Any proposed future work is subject to change based on funding levels.

21

Summary: Fundamental experiments & simulation to improve PCSI MD/HD NG engine systems

• ANL, NREL, ORNL, and SNL are collaborating to identify, understand, and simulate fundamental phenomena that limit pre-chamber spark-ignition (PCSI) system efficiency for MD/HD natural gas engines

• The project uses simulations and coordinated experiments to connect bench-scale and single-cylinder facilities to practical multi-cylinder engine and emissions-control hardware

• To extend the lean/EGR dilution limits and/or shorten the burn duration, modes of jet-ignition and resulting progression of main-chamber combustion must be better understood and then predicted through simulation

• To reduce emissions-control constraints on engine operating conditions, factors controlling methane oxidation must be better understood and new approaches must be developed to extend the low-temperature limits of catalysts

• Initial results have pointed toward unexpected in-cylinder jet-to-jet variability, certain inadequacies of state-of-the art models, and encouraging directions for new methane oxidation catalysts

www.nrel.gov

Publication Number

Thank You

This work was authored in part by the National Renewable Energy Laboratory, operated by Alliance for Sustainable Energy, LLC, for the U.S. Department of Energy (DOE) under Contract No. DE-AC36-08GO28308. Funding provided by U.S. Department of Energy Office of Energy Efficiency and Renewable Energy Vehicle Technologies Office. The views expressed in the article do not necessarily represent the views of the DOE or the U.S. Government. The U.S. Government retains and the publisher, by accepting the article for publication, acknowledges that the U.S. Government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this work, or allow others to do so, for U.S. Government purposes.

Sandia National Laboratories is a multi-mission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

This project is a collaboration between ANL, NREL, ORNL, and SNL. The project team members wish to thank Kevin Stork and DOE Vehicle Technologies Office for support of this research.

Oak Ridge National Laboratory is operated by UT-Battelle for the U.S. Department of Energy under contract DE-AC05-000R22725.

Argonne National Laboratory is a U.S. Department of Energy Office of Science laboratory, operated by UChicago Argonne, LLC under Contract No. DE-AC02-06CH11357.

Technical Back-Up Slides

24

Experimental Parameters Across Labs

ANL NREL ORNL SNLEngine/Main-Chamber DD13

No. cylinders 1 (chamber) 6 1Displacement [l/cyl] 1.85 0.38 (chamber vol.) 2.13 2.34Bore [mm] 130 100 (chamber diam.) 132 140Squish height [mm] 0.8 40 (chamber length) 5.5Bowl diameter [mm] 100 none (conical P/T and T/C port) 100Bowl depth [mm] 26 none (conical P/T and T/C port) 15.5

Main-chamber fueling Fumigation Mass Flow Controller Fumigation and/or DI Fumigation and/or DIInitial Pre-Chamber MAHLE

Volume [cc] 4.67 4.5 1.5 – 6.0 5.5 - 8.4Number of holes 8 8 6,8 8Hole diameter [mm] 1.6 1.6 1.2-1.5 1.6Included angle [degrees] 130 130 120 130Pre-chamber fueling Check-valve DI Bosch (Ford GDI part) DI DI Bosch (Ford GDI part)Spark plug M8 - NGK ER9EHIX NGK ER9EHIX (Mini Rimfire can

also be packaged)M8 – Denso/NGK Mini Rimfire

Ignition Continuous Cycles are user selectable Continuous Skip-fireMeasurements

Exhaust emissions Full exhaust gas analysisNOX, THC, CH4, CO, CO2, O2, and

CO2(EGR)

NOx, CH4, CO2,CO, O2 + detailed HC speciation

5 Gas Bench + CH4 + FTIR (exhaust/ EGR) + advanced speciation /PM as needed

None

Chamber pressures Main and Pre Main and Pre Main and Pre Main and PreIndicated efficiency Yes NA Yes – all 6 cyl YesBrake efficiency Limited (1-cyl setup) NA Yes NoOptical diagnostics AVL VisioScope possible with

cylinder head modificationNo No Yes

25

Preliminary evaluation of combustion models

Model Tunable constant Description

MZ-WSR Reaction multiplier (RM) Modify the chemical reaction rate

GEQN 𝑏𝑏1, 𝑏𝑏3 Large scale turbulence, 𝑠𝑠𝑡𝑡 − 𝑠𝑠𝐿𝐿~𝑏𝑏1𝑢𝑢𝑢

Small scale turbulence, 𝑠𝑠𝑡𝑡−𝑠𝑠𝐿𝐿𝑠𝑠𝐿𝐿

~𝑏𝑏3𝐷𝐷𝑡𝑡𝐷𝐷

1/2

ECFM 𝛼𝛼,𝛽𝛽 Flame stretch as source, P = α𝐾𝐾𝑡𝑡𝑓𝑓(𝑢𝑢′, 𝑙𝑙, 𝑠𝑠𝐿𝐿 , 𝑙𝑙𝑓𝑓)

Flame destruction as sink, 𝐷𝐷 = 𝛽𝛽𝑠𝑠𝐿𝐿Σ2

1− ̃𝑐𝑐

• Conventional model tuning was not effective for PCSI simulation

– Experimental data for all cycles showed a good correlation between timings for 𝑝𝑝_(𝑚𝑚𝑎𝑎𝑥𝑥, 𝑃𝑃𝐶𝐶) and 𝑝𝑝_(𝑚𝑚𝑎𝑎𝑥𝑥, 𝑀𝑀𝐶𝐶), which implies later jet exit will have lower peak pressure.

– However, despite typical model tuning efforts, all models kept showing an offset to experiments.

– G-Equation model offered the opportunity to tune the small scale turbulence constant.

fundamental advancements in pre-chamber spark ignition … zigler_2019_o_4.30_4.02pm_jl...chamber...

Documents